Yes. And I think this question is not well understood by anyone and is what causes the complexity to exceed what can be adequately described by anything but the most exhaustive catalog of interactions.
Essentially to my understanding, the two alleles at a locus are each contributed by one parent and each contain instructions for the cell to form a protein at a certain rate in a certain structure (which can vary to some degree). While these can be very simple, such as the classic punnet square dominance case, they can also not be simple and be additive (or perhaps another function....and in terms of frequency and maybe structure, idk...the mathematical modelling translation obviously falls into another dimension with structure). A simple version (complete dominance) is essentially when one allele is enough to lock in the protein production structure/frequency it encodes for and an extra allele with that same code does nothing new. Again, to my understanding.
The two alleles present can mutate by genetic drift, natural selection, and so on. We have identified specific alleles for specific traits enough to know this is how it works, and we have accounted for the number of possible alleles for a certain locus for various traits in humans and I am sure many other organisms.
That said, I imagine the mutation rate would be higher in plants given their general attitudes toward speciation etc so I imagine the number of possible alleles may be higher for plants. That is just a suspicion though.
Anyway, and again this is to my understanding, the effect of these alleles on the RNA and so on is far from the simple punnet square permutations that we might imagine crystallizing out from a basic high school biology understanding. In my mind, and again this is probably a pathetic oversimplification, but I like to think of it as a function that includes both a structure and a rate to then spit out the actual protein the cells create (structure) at the actual rate they create them. This actual creation of the protein can be determined by the genetics at multiple loci, i think.
But idk. I am trying.