>general relativity has had considerable predictive success, so replacing it with something else that works the same in some situations but differently in other

>general relativity has had considerable predictive success, so replacing it with something else that works the same in some situations but differently in others seems more complicated than "there's mass here in some form we don't understand yet", though people have of course tried to find such replacements. Do the models use GR? My understanding is that they make the assumption that Newtoninan gravity models are sufficient.