What's the best workflow for an animated VR 'story'?

I am attempting to make a linear piece of content to be experienced in VR. Similar in duration and content to a traditional short animated movie but viewed on a VR headset. A similar piece of content is ‘Gloomy Eyes’ Gloomy Eyes: Behind The Scenes | Oculus Quest + Rift Platform - YouTube which uses six-degrees-of-freedom.

There are many technical issues I’m running into so I’m hoping to solve some of them here. Your help will be much appreciated.

My approach so far:

I have created a main project with a ‘persistent level’ as the place to pull together the various scenes/levels and animations.

For the various shots/scenes in the ‘movie’ I have created a separate project. Once those scenes are ready to go into the movie I then import them as levels into my main project.

I am using ‘visibility tracks’ in the persistent level’s sequencer to control the visibility of those various scenes (I call them sub-levels).

All meshes (props and characters in the movie) are animated using key-frames in the ‘level sequencer’ in the main persistent level.

Questions:
Is there a better approach than the one I’ve outlined above? How long is a piece of string?

Some of the sub-levels are quite big so there can be a lag/glitch when they load using the visibility tracks in sequencer. Can I use level streaming to pre-load those sub-levels?

What you’re describing is a pretty typical workflow. Here’s some documentation and a link the to a white paper that describes more in depth about how Fortnite trailers/cinematics are structured: