Using two different powders with different burn rates means you're obviously thinking along these lines. Instead of making your powder charge "more progressive" (i.e. reserving some of the burn for later in the cycle), what typically happens is that you end up with an overall burn profile that is basically averaged between the two sub-charges. This is because you have no way of preventing the quick-burning powder from starting to burn at the beginning of the ignition cycle, so it is also contributing to the pressure curve rise (where you really aren't looking for help). If you could find a way to delay ignition of the faster powder until the pressure peak had been reached, that would be a good contribution in the direction that you are looking,
Ain't interior ballistics fun...? ;-)
Well, here's the thing about "powder burn rate." It isn't really a measure of one thing. Powders have an ignition energy and a combustion rate constant. Slow powders typically have a high ignition energy and a low combustion constant - and vice versa, but a powder that is slower than another may have a lower ignition energy. A good rule of thumb is that temperature sensitive powders have relatively low ignition energies.
Most powders produce relatively the same gas mass per propellent weight, one exception being certain shot shell powders and fast pistol powders that produce less volume due to the high ratio of nitroglycerine content. Another exception would be very slow machine gun powders that contain very high amounts (relatively) of free carbon, designed to cool the flame and slow the consumption rate.
The ignition energy of any powder is so low that, in a typically designed cartridge, by the time the first few percent of a powder charge has been fully consumed, the full charge has been ignited. Compressed loads in long narrow cartridges would provide an exception. At this point, the bullet has barely moved down the barrel and in some cases hasn't even cleared the throat. The peak pressure normally occurs shortly after the full charge has been ignited. This is due in large part to the rate of combustion of propellent being tied to chamber pressure and temperature. The combustion rate constant is a scalar factor in the rate of combustion. Temperature and pressure are also multiplicitive factors in the combustion rate.
The rate of ignition of propellent is also tied to enthalpy within the chamber, which, in layman's terms is a product of pressure and temperature. So, as the propellent burns, it also provides the energy which ignites further powder _more_rapidly. This is part of the diminishing returns on case capacity. As you add more powder, any additional charge is ignited much more rapidly than the initial charge ignited closer to the case head. And so, cartridges with a large case capacity must use a 'slower' powder with a higher ignition energy and slower consumption rate to compensate.
If you mix a fast burning powder with a slow burning powder, what you end up with is a charge where the fast burning powder will promote quicker ignition of the slow burning powder - IE, the maximum chamber pressure will come on sooner. This is the idea behind multiplex muzzleloader charges. The fast burning BP ignites a small amount of high-gas volume smokeless and, as a result, the same gas volume is produced at a lower powder charge and will less resultant fouling. Effectively, by the fast burning powder providing more activation energy, you are doing the same thing as LOWERING the ignition energy of the slower burning secondary charge. If the secondary charge is already approaching critical pressure before the addition of a fast burning primary ignition source, this is a BAD THING (tm).
One idea I've been playing around with is the idea of using a compressed charge pellet - similar to the pellets that were used in experimenting with caseless ammunition. These pellets are porous rather than a large homogenous grain and so, burn much faster than a single large grain would. A fast, low gas mass propellent could be used to evenly ignite the slow burning pellet as well as provide additional gas volume by bumping the bullet down the bore. This is how pyrotechnic shells are designed.
In pyrotechnics, we regularly layer various propellents in a shell to produce various effects, such as color change, burn rate, and fragmentation in arial shells. A similar pelletized charge could be easily enough produced for a cartridge rifle, but the caveat is that they need to get into the cartridge behind a bottleneck. I had an idea that one could make a compressed propellent "stick" to be loaded down the bottleneck as a secondary charge and then an additional granulated primary charge could be added around the primary charge. I tried modelling these and realized it wouldn't work because the secondary charge would lead to inconsistent ignition of the primary.
What I came up with is the idea of using a stick-type primary charge. A very very fast shotgun powder is compressed and extruded as a single tube. This tube is then surrounded by an extruded grain type very slow secondary charge with a very high activation energy (like a modified machine gun powder). The flame front from the primer will ignite the primary charge very quickly from the head to the shoulder and also provide additional accelleration for the projectile, reduce the internal chamber volume and pressure. The secondary charge is ignited by the primary charge, but due to the decreasing volume and delayed ignition, reaches peak accelleration much later in the game. The gotcha is designing the compressed primary charge such that it sits consistently in the case throughout the combustion. A stick form is ideal for this since it could be designed to anchor in the primer pocket.
It is also ideal to create a primary charge that burns very fast near the head and slower near the throat to provide additional activation energy to the TOP of the secondary charge. These charges could be produced in a press as a cone narrow shape with additional carbon added near the base to slow the burn rate there as well as shroud the primary charge near the base.
For obvious reasons, the barrel would need to be designed very strong for quite a distance up the barrel. Alas, I don't have the technology to test this safely, so I have not.
I'm also not sure what the legalities are of testing something like this, but, in a home made canon where "anything goes", would be fun to try.