Hannes pretty much answers your question.
To the best of my knowledge the sequence loading functionality works exactly like the normal image or object loading system, the only difference is it knows to use a different object file for each frame, and it gets the info on which file to use from the file name wild card specified, as in Hannes' example above. So basically, you start rendering with frame 1, it loads birds0001.obj and renders the scene, frame 1 finishes and saves to disk, frame 2 is about to be rendered and now the OBJ loader sees it's on frame 2, it loads birds0002.obj and renders, and so on. So as long as a single instance of your model loads and renders fine in TG, the sequence *should* load and render correctly. After all it's literally just a series of separate OBJ files, no different to if you manually loaded a different OBJ and pressed Render a bunch of times.
The only issue I can think of is if the program that is outputting the OBJ doesn't output multiple corresponding MTL files and only puts out 1, but hopefully that's not something any of them do.
What you need to keep in mind is that fully animating complex models is going to create a lot of very large object data. If you have a 10MB object and you want to animate it over 1000 frames, you suddenly have 10GB of object data. It could easily get into 100s of GBs very quickly.
- Oshyan