Overall I am saying that in an environment which is not rapidly changing, species would evolve more accurate means of copying DNA
I don't see that this is likely. It sounds like you are suggesting selection at the population level.
The only way this could be selected for is if having fewer mutations increased the rate of live births. It seems to me that other solutions exist for this so the lower mutation rate may not be encountered and selected for over those.
As was suggested above, if almost all mutations are harmful (because a population is very well adapted to an unchanging environment) then the usual natural selection would remove them. The reduced mutation rate would have to, somehow, result in greater fecundity. That may be possible but might not be reached.