Many-worlds and Copenhagen are the two major interpretations (actually features) of quantum mechanics, and both are valid to some degree. So, instead of arguing and battling over their differences, why can’t we think of a way to bring the two together into a single mechanism. The question is, what it takes to envision a scheme that can accommodate both Many-worlds and Copenhagen?
While we are in this unifying different interpretations (or features) mode, maybe we should also throw in the Bayesian interpretation (feature) as well.
To clarify, each valid interpretation in QM actually reflects a feature in QM. That is, MWI is a distinctive feature in QM and so is Copenhagen, and that is why we say both are valid in terms of physics. People tend to use the word "interpretation" to refer to a major feature in QM and then somehow regard it as a sort of man made fantasy since it is called an interpretation. The feature referred to by a valid "interpretation" exists just as the whole QM exists.