Honestly tracks don't do what most of us who are used to SCM systems like git would prefer. There is no diff / merge capability at this time, so you can create a patch in one branch and then "merge" it into another. There really isn't a good way to use tracks for concurrent development as they are implemented today.
What they are good for is for creating a patch to production. Since you can create a track from a snapshot, if you need to create a patch for your production system you can create a track for that patch, fix and test your code, then make a new snapshot for promotion and testing. Unfortunately if the items you modified have been modified in your "main" branch as part of,say, the next release, then whatever you did in your track will need to be manually duplicated in main.
If the item you modified as your fix in main are unchanged, you can copy the modified items back to the main track (actually you can do this anyway, but you will over write the other changes done in that track)