Why complicate things?
Go with your original idea of a passive crossover. You just may need to play with the actual cut off frequencies.
You'll also want to use two different frequensies, to attempt to get a smooth transition between mid and tweeter. Usually the 6 db down point of each crossover frequency alignmnet is what is used, so that when both speakers are playing the combined output at the overlap will become a flat or close to responce. I think I might actually look at using two different slopes as well. Find the natural upper roll off of the mid and run a 6 db sloe for the low pass, and then a 12 or maybe even 18 db for the tweeter. I also prefer Butterworth slopes, in most cases, though Linkwitz riley slopes can be used to soften a tweeter that has harsh low end.
The good thing is that you could place the passive components in such a way, that you could access them easily, to play with down the road. Maybe even use a bread board so that you can quickly and easily change components to find what best suits your music, car and components.
It will still sound decent using settings that aren't completly matched, and it's always be a compromise anyway. This will be a cheaper way to go, even buying components for several different configurations will likely be cheaper than buying an extra amp and active crossover net work, the best part is that you'll learn more about passive crossover design than you wanted to know and know how to tweak the x-overs.
I've ran both active and passive systems, and honestly I can't say I hear much of a difference in the end, as far as sound quality goes. The active system plays a little louder, due to a more efficiant transfer of power from the amplifier to the speaker, but that's about the only sound difference. The active set-up usually allows fo easier tweaking, but also comes at a higher cost, especially when you get into the ability to change slopes and alignments.
