Yep... happens all the time. That is why I cold start an amp like this with an incandescent lamp in series to act as a current limiter. I will tell you what is wrong. You did NOT go back far enough in finding the bad parts. Also some that were not shorted MAY have been degraded. The real problem lies in the driver transistors and the bias regulating circuit and the transistors in that area. This amp has a mass number of components and there are many that if they fail they wipe out a bunch. When you think you have it repaired run it a while with a 150 Watt light bulb in series with the power input cord. Here is your schematic:
Scroll down to the link "Get Manual" and click to download. In the schematic there are over twenty transistors involved with power output and bias stability
. This is one of the most complex amps I have seen. They get voltage capability by having two series banks of transistors to each voltage rail. The biasing network is a nightmare!!! Pot R28 sets the rest bias which you measure between J5-1 and J5-2 which is 1/3 of the rest current across a total of .4 ohms... whatever it is supposed to be. Q3, Q7, Q16 and Q17 are ALL real critical to the biasing as well as the smaller Q12 and Q13 and all the others. The problem when you have a feedback amplifier is if any of these components are sick, the feedback will still attempt to balance the amp even though it sloowly fries other parts. Frankly, this is going to be a miserable one to fix. Use the light bulb trick and a Variac. There are a few voltages marked n the schematic USE THEM TO CHECK the operation. When certain transistors go, they will take a bunch with them... NOTICE the FUSIBLE resistors up near the rails a couple stages back... need to check all of those and replace with IDENTICAL ones if they are blown or way off value. There are a lot of Zener diodes that ALL need to be checked as they afect the bias levels. You will need nearly full voltage during checking which really makes it tough. With the light bulb if too much current is drawn from the line the light will light absorbing the power line voltage. The light should remain off when repaired AND voltages should be near what is shown on the schematic. I would expect to see half the rail voltage across each of the sets within the banks when at idle. Good luck... from an electrical engineer with 50 years exp.