If we hit an actual wall, then that could be the missing part of the unified theory, some boundary that needs to get applied to the equations of both that makes

If we hit an actual wall, then that could be the missing part of the unified theory, some boundary that needs to get applied to the equations of both that makes them explicitly inapplicable in the regimes where today we just observe they don't work. Of course, we might also hit an "engineering" wall, where we don't have any explicit boundary, but it's just too hard to perform the experiments that would help us understand how these theories relate. That would mean that the current status quo will remain the best we can do - two separate theories which we know are wrong, but where for any experiment we can perform, one or the other predicts the results almost perfectly.