Sounds like some of the vocals are now a bit low compared to the off-vocal. I think each vocal needs its own compression before mixing. The off-vocal already sounds heavily compressed to me, so the vocals are going to be hard to mix without having a similar narrow dynamic range. FL Studio has a simple compressor, which I'd recommend over the limiter/compressor if you're not familiar with it since the latter can be difficult to configure. Or you can use the compressor in Audacity on each rendered vocal before bringing them into FL. But with the latter option there's no tweaking of the compression once you save the file. That's what's so nice about the effects in FL: they all act like filters on the input rather than actually modifying the input file.