You gotta pick one at the end of the day.
I'm not building a hybrid React / Angular / Ember / JQuery app, that's not efficient. And picking one is difficult, because they each have their strengths - as well as nuances.
That's where the debates and line drawing begins.
Coders are like artists, we pick our tools and fields and gain experience in them. You may ask me to design you a package, and I can ask: "Do you want me to use a brush with oil paint? Or a digital stylus and illustration software? Or a photograph from my digital SLR?" We all get to the same end point with different methodology to achieve unique results.
And we all have limited time. I only have so much life in me to learn every language. So I'm probably going to pick a top 3, or 5, and focus on those. And then I've got to figure out how the languages meld with my team.
I'll also probably have strong opinions about my choices, because I don't make decisions lightly. And that's O.K. too, because that passion will translate into a deeper understanding of the ecosystem. It helps justify decisions when someone understands why they make them. "I don't like React for this type of function because Angular handles this better and here's an article from Evernote with data proving it." It's a healthy debate, rather than "X is better than Y period".
Part of the KISS is the initial debate of what the simplest route would be, rather than being forced to refactor opinionated code later ("Why are you using Angular for this one form when the rest of the site is Mustache? I thought we agreed on this...").
It's like you said, you just pick the best tool for the job, but not every artist is trained in every tool. I won't force a team to go Node if they've never touched it and prefer PHP.