Resistors almost never go bad to the extent of changing value as you describe. Your reduced ohm readings suggest a parallel component is practically shorted.
Did you check ohm readings of the 324 pins? Use the highest (1M) range rather than the lowest (200 ohm) range. If many pairs yield a similar low reading then its innards are no doubt fused together.
If the 324 were socketed it would be easy to say 'Replace it with new'. However to desolder it is a headache. For now test other simpler things.
A variac would come in handy right now. And an isolation transformer. As a substitute you might jumper an incandescent lamp bulb in series with your unit. 100W is about 200 ohms at 220VAC.
It sounds as though the transformer contains 2 or more windings? (4 thick wires go to it.) (The remaining two small wires most likely convey low voltage to the board.)
One spark arc event can create a shorted winding, then the whole thing stops working. And most likely overheats due to extreme Amperes going through it. Arcing can happen if shutting off high current through of a very high inductance load, or if a lightning strike sends a high voltage surge through the lines.
Eventually you may need to test each winding for inductance (Henry) and resistance values, and all parameters that might give you a hint whether any winding is bad. Also to test step-up or step-down capability.
To do it properly you must disconnect all wires to the transformer, and tag each wire. It's tedious. If you find a winding is broken (open or shorted), then you must remove all turns until you find the problem, fix it, then wind all wires back up. This is such a big job that you should first try every test possible with a meter, or lamp bulb, etc.