Yes, brand new whole master cylinder assembly, P/N:
47050-60041
Brake Booster Assy, W/Master Cylinder
W(VSC),PNC 89637D=TYPE A
(Note, I didn't order it from ToyotaPartsDeal.com, I ordered it from my local dealer for about $100 more plus tax, free overnight delivery... it helps to be close to the parts center in KC!)
Last night, I disconnected the skid control ECU (battery disconnected!) and all the connectors on the MC assembly. I cleaned all of the connectors with electrical contact cleaner. Then, I readjusted the brake pedal so that there's now about 3-4 mm of free play before the brake lights come on. This puts it at just about 7 inches above the floor, which is lower than the FSM states it should be adjusted to. I'm pretty sure the clevis is at approximately the same position as when I took the MC out of the box.
Hooked everything bag up, started the car, reset idle, and connected Techstream. I ran active test on the ABS/VSC module (which cleared the codes) and then, for s***s and grins, reset the zero point calibration. Taking the kids to school this morning, I threw no codes (yet...) So, I am hopeful that the issue is resolved. I won't be able to say for sure for at least a few days, given how intermittent it is, but knock on wood it is resolved. I'll hook up Techstream later today to see if any ABS/VSC codes are pending.
So, if it is indeed fixed, it seems like there is (was?) a conflict between what the MC pressure sensor was reading (i.e. some pressure) and the fact that the brake lights weren't necessarily on because the pedal was adjusted too high, and it wasn't depressed far enough to turn on the lights. The skid control computer seems to resolve these two signal against one another, and if they're not in sync, it's not happy. I'm assuming the stop light switch/circuit also controls the shifter interlock, because that feels normal again. I'll keep an eye on it and report back.