The corpuscular theory of light was one of those theories that was wrong but useful. The successor wave theory was also wrong. It may turn out that general relativity is founded on incorrect reasoning ("God does not play dice" for one), but that matters much less than it being fantastically useful.
The basic idea- if light interacts with gravity, if you pack enough mass into a region you will get a dark star- is correct. It just turns out that the interaction can't be described by treating light as little particles and calculating by Newton's equations.
There's been lots of wrong-but-useful theories, like phlogiston or arguably luminiferous aether:
It's not founded on incorrect reasoning, which would imply a logical inconsistency or error in the deductions, but it could be founded on false postulates. (A postulate is like an axiom except that it is held to be true in the real world, as opposed to being abstractly true.)
The basic idea- if light interacts with gravity, if you pack enough mass into a region you will get a dark star- is correct. It just turns out that the interaction can't be described by treating light as little particles and calculating by Newton's equations.
There's been lots of wrong-but-useful theories, like phlogiston or arguably luminiferous aether:
https://thonyc.wordpress.com/2015/10/23/the-phlogiston-theor...