Yes, the example I gave of jellybeans and chalkboard scraping is quite simplified in order to make the point. An understanding of the complex interplay of creating expectancies and violating them is an understanding of the history of musical style. I'd say that expectancies can be created both through years of conditioning, resulting in an overall style, or even era; and they can also be created in the short term, which is why we can readily adapt to individual composers' styles. You do seem interested in thinking about this, and I'd say Leonard B. Meyer's books will be useful to you at some point. You could try and see if you can understand his writing at this point, but if not, you will probably find them very useful in the future.
You will likely find a study of theory and harmony to be useful. It's not exactly true, at least in classical harmony, that an accidental is used for surprise. In classical harmony, notes are used in a way that corresponds to a very intricate set of rules of how to approach them, how to leave them, and which notes can be used under which circumstances. In Bach's harmony, notes ALWAYS have a reason for existing and are never introduced willy-nilly for shock value. Part of Bach's genius was in how he was able to stretch these rules to allow some very unusual harmonies at times.
Leonard B. Meyer talks about musical composition as involving certain strategies and decisions made within those strategies. What you described is a possible strategy for composing a piece of music, but just be aware that it is not one that fits with classical harmony. No problem, though. Part of the creative process could involve creating your own style. But much thought, experimentation, and experience are needed to make such a newly invented style work.
Your argument concerning tuning is an interesting one. But before going into it, I want to make sure you understand what I was writing about. There are two choices to be made: one is the pitch, and the other is the temperament. The pitch is what I think you're talking about. How high is middle A? Is it 440, or higher, or lower? The second, temperament, concerns the relationships between the notes. If you tune two different temperaments pitched at A=440, for example, the note E flat might be slightly different between the two temperaments, so in one temperament the key of E flat will sound more in tune than in the other one. If you want to know more about temperament, just ask and I'll try to explain. It's a fascinating subject.
Anyhow, I believe temperament is a more important issue than pitch, though there are some very important differences that pitch can bring to singing, especially. A good composer writes vocal music with the singer's register changes in mind, and no matter where the music is pitched, the breaks occur in the same place. So if, for instance, a Baroque composer wrote a piece for a tenor with a high F sharp, thinking it will be at the top of his chest register, the singer will have to sing it in head voice if the music is pitched at modern pitch. That could be a pretty significant problem, and actually IS a problem with some of the vocal solos in Handel's Messiah, for instance, if it's sung at modern pitch. There are some other, more subtle reasons, too.
But to me, the differences in temperament are even more outstanding. To play an early piece intended for Meantone temperament in modern equal temperament makes it sound profoundly different. And it is very important in my opinion to let go of the idea that modern is better. It's just different when it comes to art.
There are certainly preferences regarding temperament among modern musicians, and many prefer equal temperament. But I know of very few who actually like equal temperament on a harpsichord. This is not simply a subjective decision, because the richer overtones of the harpsichord make equal temperament clash more than it does on the piano. In equal temperament, none of the intervals are purely in tune, and this becomes obvious on the harpsichord. However, WHICH temperament you use is a subjective decision. I prefer Werckmeister, and Meantone for earlier pieces. The beauty of a harpsichord is that once you're experienced with tuning it, it's not such a big deal to comnpletely change the temperament in order to accomodate a different style or a particular group of composers.
Keep thinking about these things, because you ask good questions and you make good observations!