Not one but *two* fabulous disclaimers for this one:
1) You do the hobby the way you want, regardless of what anyone else says
2) I fly Estes motors at probably 80%-90% of the time because they suit my needs well in most instances.
I don't understand the reasoning for arbitrarily limiting yourself this way. Composite motors offer very different thrust curves than BP motors that open up new possibilities. Why avoid them?
Let's go through this point by point:
Yes, no changes required.
I don't know if I've ever heard anyone complain that a composite motor was less exciting than a BP motor. But ultimately there is exactly one way to answer this particular question, and I think you can guess what it is.
They are designed to fly in the same rockets. I have never done any specific mods to accommodate composites, with one possible exception (discussed below). To be fair, ejection charge is always a bit variable, and is something evaluate when you use them.
Define "expensive". Estes E12s at maximum discount
(i.e. ACSupply) are about $5 a piece. Q-jet E26 are $7.50 each from Sirius. A bit more expensive? Yes. Break-the-bank-don't-ever-use-them expensive? Not from where I stand. The Aerotech E20 and E30 cost a bit more, but if you need the higher total impulse (or in the case of the E30, thrust) then you get what you pay for. The E20 is one of my favorite motors, for sure, although I don't get to fly it very often.
The 18mm Q-jet D16 ($5 each at Sirius) is great for 18mm models that need a bit more umph than you can get with BP motors, and I've also used them in 24mm models with adapter when I want less altitude than a D12.
The one exception I mentioned regarding "build to accommodate composites" is that they are usually, at least in the single-use varieties, somewhat higher thrust than BP motors. That's great for getting heavy rockets moving off the rod, but in many cases this will also subject the model to higher G-forces, so your builds must be sufficiently sturdy. My eyes were opened to this when I recently saw @rklapp
's in-flight video of a Super Mars Snooper flying on a Q-jet (D22 or E26, don't remember which), and seeing the fins flex at the initial boost. I haven't yet made any special accommodations for this, but I do tend to overbuild a bit to begin with, and thus far I haven't had any issues. But I haven't flown an E30 yet (I have some but the opportunity hasn't arisen). It is just something to keep in mind.
Simulating the rockets, either with OpenRocket or just using thrustcurve.org, will give you an idea of the rod speed and G forces you'll be experiencing, and will allow you to plan accordingly. Estimating delay time for many of your rockets will be more difficult, because your rockets tend to have unusual features that really go outside the bounds of what the simulators can handle. But you can get a feel for things after a while.
Ultimately, my one real complaint with the 24mm Q-jets and single-use Aerotech motors is that the delays are offered in 4/7/10 seconds, and so many of my rockets seem to want 5 or 6 seconds. But a one second error is not a dealbreaker.
So, bottom line: as disclaimed above, enjoy the hobby the way you want to. But consider taking advantage of the additional options that are out there.
covered most of the same ground in about 1/4 the words]