I like establishing a specific path, outside of JRiver's installation directories, to hold additional libraries. That way I know they are entirely separate and I can back them up, move them, etc. Installation, reinstallation, updates, deinstallation, etc will never touch these directories.
I generally do this for experimentation purposes, or when I'm working on other people's libraries for various reasons.
Your "extra" libraries will be backed up *when* you use them. At some point, after a couple of days of not being backed up, MC decides to run an automatic backup. I'm not sure what triggers it. But I am reasonably certain that to back up library X, you must have library X open in MC. To have library Y get backed up, you have to have it open in MC. Which makes sense. Why would it back up a library that isn't open and therefore hasn't changed?
As the guys said, generally speaking, you can do what you want with a single library. This is usually more efficient and convenient. There are some instances when extra libraries make sense. But they are the exception, not the rule.
Good luck.
Brian.