Part of it is about how quickly the heat transfer must take place. If you were relying on convection to transfer heat into the air, aluminum works pretty well. But there isn't enough time for convection to happen with a clutch, not to mention the clutch is in a confined space. Add to that with an aluminum flywheel you are trying to transfer heat from the insert to the aluminum. There are small gaps and spaces between the two materials which act as insulators to inhibit heat transfer. Remember the heat conductive gel used when mounting some electrical components? It's used to fill those gaps and spaces to improve heat transfer between two surfaces, but that gel is not really practical for a flywheel/insert application.
Then there's the insert being attached to the aluminum, with the aluminum providing most of the structure to keep the insert flat at room temperature. When slipping, the insert absorbs heat faster than it can transfer it to the aluminum. Because the insert has little mass it's temp rises very quickly, causing it to grow at a faster rate than the aluminum. When this happens, the areas of the insert that are bolted to the aluminum are restrained somewhat compared to the areas that are not. This is causes the insert to warp, and with that warping comes loss of intimate contact between the two materials which is needed for direct heat transfer. Pretty much the point of no return as not only does the insert temps skyrocket at this point, but so do the disc temps.
The friction materials used in a clutch are basically the same as those used in a brake system. As with a brake system, they will all last a long time as long as temps are kept in check. It's the reason an oval or road course car with heavier rotors can post quicker lap times when that course requires a significant amount of braking. There's a reason you don't see steel inserts on aluminum rotors.
A flywheel is an energy storage device. The entire rotating assy is the actual effective flywheel, the part we call a flywheel is only one part of that. When you spin that flywheel up, you are charging it with energy. That energy contained is a function of the flywheel's rpm. The more energy you charge it with, the faster it will spin. And when a flywheel loses rpm, it's because energy exited that flywheel.
You can quantify the energy contained in a flywheel. Lets say a given flywheel spinning at 5000rpm contained enough energy to supply a 500ftlb boost of torque for 0.25 seconds. The clutch is what determines the rate that you draw that stored energy out of the flywheel. The clutch could draw out 500ftlbs for 0.25sec, or the clutch could draw out 250ftlbs for 0.5sec. Both are the same amount of energy that gets dumped into the chassis, 500x0.25=125ftlb/sec or 250x0.5=125ftlb/sec. The difference is that an additional 250ftlbs over 0.5sec is a hell of a lot easier for the chassis to process than 500ftlbs over 0.25sec. The CT allows you to dial in the rate that the clutch pulls energy out of the effective flywheel. Rpm isn't a factor in determining the flywheel's discharge rate like it is with your adjustable clutch, which allows CT users to launch off the high side and dead hook as long as they have enough clutch.
The starting line advantage isn't flywheel weight as any added inertia boost during launch due to added weight has to be paid back as slower acceleration before you reach the finish line. The added weight thing cancels itself out in the end. The advantage comes when you can process a higher staging rpm while dead hooking without damaging the clutch. The flywheel is going to be fully charged when you cross the finish line, so any additional power you have to spend spinning up the flywheel is power being diverted from accelerating the car. Power in basically equals power out, so why start the race with the flywheel half charged when you no longer have to with the CT?
Grant