I also realize that there are some parts in the story that doesn't fit or work and I need to figure out what's right for the path here instead of trying to push forward.
Generally, it's best to write while the proverbial iron is hot. Get it out on paper, and then you can reevaluate what 'fits' and what are mere distractions during the revision process.
I've deleted whole chapters, subplots and entire characters from stories, though many here instead do their revisions on a chapter-by-chapter basis (before they know how the entire story plays out). However, even in those cases, they'll typically revise select passages when they realize it doesn't aid the story.
Also, if the story runs long, you can also break it into multiple books. You'll need to be careful, as it's generally better to select unique story conflicts and story objectives (what the story tells the reader), but you're at an excellent point to evaluate that now, before you get any farther into the story which might require backtracking if you change your approach in a subsequent book.
My objection about characters 'taking on a life of their own' is that you need to understand what's at stake and whether it's related to your approach to the story. However, once you've evaluated it and decided how best to resolve it, it's often easy to correct. Often, simply acknowledging an issue in a story is enough to buy you time to develop the story with your readers. They'll know whether something doesn't 'ring true', so letting them know it's a part of the story will get their buy-in and they'll wait to see how it plays out before passing judgment on the story. (That's why beta-readers are so valuable, as they'll often flag story issues you never conceived of.)
From the sounds of it, it sounds like you're on the right track. You're aware of issues and are already considering alternatives, instead of just bullying your way through the story (which is when writer's block truly runs rampant).
The one thing I would caution, though, is starting too many competing stories (was that you, or someone else?) While Ernest does that often, having too many concurrent stories tends to sap your attention and focus, causing you to miss essential issues with a story.
It makes sense to work on multiple stories at a time. That way you can write one while revising or editing another (corrected your editor's corrections), thus keeping your skills fresh. But having too many failing stories is a definite sign it's time to table a few of them. :(