I wouldn't exactly say I've released any games (I've made a couple of hobbyist games a long time ago though), but...
If you've already reinvented the wheel, then why adapt it to a formerly invented one?
For one, if you are planning on having this stuff in a portfolio, then having done an engine is nice to show off.
On the other hand the jMonkeyEngine looks great...
So, if you really need the features it has (and it doesn't actually sound like you do, it just sounds like you want to use it to add lots of extra stuff to your game), then it might be a good idea, if not, you're already almost done.
If I were you I would probably finish off the engine I was working, then for later projects, I would reconsider using the jMonkeyEngine before starting off.
Really, having done an engine from scratch is a lot of work, and is worth showing off, even if it's only used for a single project.
It's a learning experience, and you were going to port a lot of the stuff over to jme anyway, which you should do in the future, when you start working on your next project.
It's important not to decide to add a bunch of features halfway through making a game, as then when you're halfway through adding all that stuff, you'll find a bunch of new stuff to add...