Multiple Nested Timelines or One Timeline for 40 minute Short Film?

Great question... I can't help with that. Never edited anything that long, but am hopeful you'll get an answer so I can read it too :).

I've done a few long form edits...a couple of 3 hour operas, a big band concert, artist interviews...and have found there are some real advantages to editing in blocks which are usually, for me, on separate timelines or, in the case of the operas, separate projects...at the end of the edit I bring the separate sequences, either nested or grouped into one timeline for exporting.
The advantages in no particular order for me have been:
* working in segments that are on separate sequence timelines allows faster navigation through each timeline...purely because they don't take up as much real estate on the monitor...this can speed your workflow
* if you group or nest each one it speeds up the sequencing on the main timeline - you don't have any missed bits (transitions, added small audio bites, very short clips or stills/graphics) if you move the segments around to get the optimal sequencing
* you can colour grade each segment and ensure all the clips for that segment are matched more easily
* it seems to be more efficient in terms of computer speeds - I'm editing on an old machine...this is why I divided the operas up into acts and had a separate project for each...my computer just couldn't handle 3 hour timelines
hope this helps...I'm sure others will have other workflow comments to make

I agree with John Cliff. Breaking down the entire video into smaller segments helps a lot. In case one wants to change the flow of the story, it is just to play with the sequences rather than playing with the clips. Of course, one will need to fine tune after that.
But there will be certain unique situations where you will need to figure out the way. Sometimes, the background score or audio continues from one sequence to the next, there I use the main time line to synch those audio, rather than in the sequence.
Try it once Christopher, you will enjoy it. Though it could be a difficult beginning. :-)

Did a feature (1h40) this way years ago. As I wasn't as adept at editing larger projects then, breaking things up into nested sequences, sometimes down to the scene level, was really helpful. It helped me edit in chunks and then move those chunks around if I wanted to work out the order, etc, without accidently disrupting any of the video or audio layers, etc.

HOWEVER, I would also advice that once you are done with the edit, or mostly done, you move everything into one single timeline. The main reason for this is audio. If you use nested sequences, depending on how you nest or how you combine those nests, you'll end up with a single stereo audio output. Meaning if you have two lavs, some room tone, and music in a nest, these will only output as a single stereo left/right channel. That means you can forget proper sound mixing, etc.

So, do whatever feels right and make sense in your head, just remember to keep the audio tracks organized across all nests and then combine the edit in the end.

I finished a feature (78 minutes) recently and I learned quite a bit along the way. Source footage was UHD from two Sony mirrorless cameras. I first created a sequence from each scene and edited them accordingly. Then I copied and pasted into sequences that consisted of several scenes, and did a pretty extensive color pass. The real time-saver was that I set up the scene group sequences in DNxHR HQ 2K, which meant that the preview file format and codec match the output format. So after rendering, I exported .mxf files of the scene groups and put them (sans audio*) in a layer on top of the files I just edited. The effect is that there is NO rendering required after this point. The line across the top of the sequence isn't green, it isn't yellow, THERE ISN'T ONE. Next time you export, it only takes as long as it takes to write the file to your drive.
* There's a bug in Premiere that when you export .mxfs, regardless of the export settings, each audio track is a mixdown of the project as a whole. So deal with the audio separately, then do a final sequence where you line up your group .mxfs and drop your final audio files under them.