Again I have to confess to perhaps missing the point - you mention Everett but it seems like you are bringing the observer into wave-function collapse? And I thought MWI removed the observer?

It's more that I don't see branching, the many worlds thing, as particularly relevant in the context of universal wavefunction. The function itself is interesting, as being all-encompassing in a not really clear way. In essence, there is one "wavenction", whatever that might really be, that is the whole universe. And that function is always the same, while containing all the infinite collapses within itself. And yes, it would absolutely be valid to interpret those collapses as the universal wavefunction interfering with itself.

I say all of this as a counterargument to your not seeing relations without relata. In this view, there is precisely 1 relatum, the universal wavefunction.

I'm pretty sure that MWI does not "remove" the observer. If I understand it correctly, and this is a big if, all it really says is that any time the wave function collapses, be it from observation, or just from, say, radioactive decay, there is a "branching."

This is correct, MWI doesn't really concern itself with the observer effect, it's not interested in why the collapse happens, it wants to explain what the collapse is, or how it should be viewed. This is exactly why it's not really popular, since its explanation is not constructive.