For the past two weeks we have been discussing how to support the scenarios you need to make camera stacking, UI and post-processing.
Rethinking about the problems it all boils down to the fact that we need to allow to run post-processing stack at different parts of the camera stack.
Allowing to run multiple post-processing stacks on the camera stack is not trivial feat, but it seems running post-processing once and allowing to chose where in the camera stack to run it would cover majority of scenarios, including the world space UI above. Depth can already be optionally preserved with Overlay cameras, so the UI carrousel above should work in the new version. Also once custom post-processing is supported you will also be able to inject any of these in any part for the stack as they are essentially a render pass.
So we are prototyping this solution: expose in the stack a post-processing event you can choose where to inject.
I would like to ask to keep this thread focused on post-processing and move the camera stack feedback here: Camera stacking for URP