Your argument is correct based on your premise. That is exactly what torque curve means. It is a thorough deion on output of an engine, pretty much everything you need to know. Saying only max power determines acceleration is wrong, at least in the real world. One point on the curve can not determine the entire range.
Is max power still a useful concept? I think it is. Now you need to think beyond math. First, max torque and max power give a first degree approximation of engine performance. You can roughly draw the torque curve in no time. The curve would creep up to max torque, then go down to Pmax/(it's rpm) then plummet from there. How many times when you look at a car, you take a look at max torque and max power? For me, almost 100%. How many times do you really collect torque curve and gear ratios and calculate 0-60, 0-100, 0-120? For me, it is almost none. If you are like me, you already admit max power is a useful concept. But if you do that kind of calculation most of the time, I do not know what to say (actually, I wan to say you are nuts).
Secondly, your premise is that torque curve and gear ratio are given to you. So you pretty much just sit in a car and drive it. Original buick vs Toyota discussion sounds like discussion on low end torque vs. high end torque/power. (I did not looked at those posts). This is beyond acceleration. Even if you only care about acceleration, you only proved torque determines acceleration. You did not prove which one is better. To answer that, you need to look from engine design/tuning point of view. You need to see why torque curve is like that, how you want it to be, instead of given a fixed one. Looking from this angle, you probably would find max power a concept you can not do without.