The only thing I think you would gain is a little wider stereo sound field on your lead. Having said that, you could get the same width by panning your #3 and #4 tracks a little further to their respective sides.
There are a lot of reasons to do exactly what you did in this track. The only time you can get it wrong is when your mix sounds like crap to your own ears. Otherwise, if you like what you're doing, both playing and mixing, it's all good. The splitter won't do anything that you can't get by other means in the DAW.
Also, after doing a little more reading and watching a few more videos, it looks like normalize always "normalizes" to 0db, regardless of where the mixer fader is set. I was fooling around with some macros in Studio One and one of them was 'Normalize -6db.' In looking at the construction of the macro, it appears to be done in two steps, with step 1 normalizing and step 2 dropping the event by 6db. I looked all over the macro tables and a bunch of other docs, including Presonus forums and it appears the target for all normalization is 0db in S1.
Honestly, I hadn't been using normalize with S1 until I saw that Griff is such an advocate of using it. I didn't even know where it was in the menu. But there is certainly no harm in using it to tame a backing track before you record a contribution to it.