carpa, for going on forty years I've recorded classical music almost exclusively, including quite a lot of solo piano. In the early 1980s I was the engineer for a two-year series of weekly solo piano recital broadcasts a classical radio station here in New York, for example. I was also the digital technician on several CDs with Ivan Moravec and two of the late Paul Jacobs' recordings for Nonesuch (and in part, was the uncredited engineer/producer for his duo-piano recordings with Ursula Oppens, which I believe was his last recording).
I make my living these days doing technical work for a publisher, but I still do a fair amount of recording as a volunteer, to help performers and organizations I believe in and just because I enjoy it. Even though this Web site isn't a classical-oriented place, I enjoy it here because of the intense mixture of rowdiness and caring a lot about music, sound and life itself. Most of the people here are far less cynical than a lot of people who are professionals in the audio business. I need that.
Anyway, we could talk, and I'd be just as glad to do that here in writing and in front of everybody, if that wouldn't be too exhibitionistic. I see you trying to work out a compromise between convenience of setup and sound quality, just like most of the rest of us (I'm past the days when I would lug a 40-pound Studer open-reel tape recorder and two heavy mike stands to every concert, though I did that for a while).
The thing is, when you're recording yourself playing, you can't be trying five different setups and comparing and deciding before you perform; there will be times when you'll listen to the playback and realize that it would have been better if the mikes had been a foot closer or higher, or if the levels had been set 6 dB differently. But it's better to have those regrets than to have a perfect recording that only gives you regrets about how you played that night.
I don't know how particular you are about the type of sound you'll get from different types of setups. There are fundamental differences between (to pick two extremes) a spaced pair of omnidirectional microphones versus a coincident pair of strongly directional microphones such as "figure-8s." These two recording methods give results that not only sound different but feel different--one listens to them rather differently, I'm convinced. There are also some plausible midpoints between the two (e.g. closely spaced, but not coincident, moderately directional microphones).
One thing that you mention which I would probably not recommend is the Røde NT 4 stereo microphone, unless you have tried it and found that it coincides exactly with your tastes. In my view its design (coincident cardioids with only ±45 degrees between them) is suitable only for very close-up recording, or for capturing discussions at a wide angle around a table, for example. Such a narrow angle between capsules of only moderate directionality will give you a nearly mono pickup unless the sound source is unusually wide. For normal classical recording which is picked up not too close to the sound source, a considerably wider angle between capsules (e.g. ±55 - 60 degrees) would be preferable by far.
But (again, unless your ears tell you differently and you are confident about your preferences) cardioids are probably the worst possible general choice of a pattern for use in a coincident stereo recording anyway; they're fine for spot miking in studios or live work, but semi-distant stereo pickup is a whole other thing.
--best regards