A couple of things to note. Every open bracket has an appropriate closing bracket. This would indicate they have a specific meaning, rather than having their values picked by a polyalphabetical cipher. The use of brackets is heavy in the first note but lacking in the second note. The brackets are also never nested. The space between them is sometimes too short for them to be deliminating individual sentences, and they are not used very much in the second note. It is possible that they represent quotations. They could actually be used to mark regions which are using different cipher keys whether they be monoalphabetic or polyalphabetic and that would make looking at the frequency of letters a bit more difficult, but I don't think that is the case because we do actually see repeated patterns across brackets.
This is not merely a transposition cipher. If it were, we would see a much higher occurrence of the different vowels. The only one that has frequent occurrence is the letter E. That doesn't rule out the possibility of it being a combined transposition and caeser cipher.
We can see certain repeated patterns. In particular many of these words seem to end in the letter E, and we have some repeated patterns.
If you look at the lines on the bottom note, we see:
and similar patterns occur earlier in the same note. WLDNCBE appears on lines 5 and 6. WLDNCBE also occurs on the third line of the second note.
We can see other patterns which have a similar form to them. ERTE, FRTSE, NARSE, PSESHLE, VLSE, MTSE, CTSE, GLSE, all of these appear at the end of a word of sentence, and such forms do not appear to be at the beginning. The beginning of the words seem to have less recurring patterns, but there still are a few. In particular, a lot of them are prefixed by TF, TE, or AL.
Looking at this I think this isn't a transposition cipher, polyalphabetical cipher, or monoalphabetic cipher. But because of this repetition I think he is nonetheless using a similar ciphering technique. It is possible that he is actually ciphering entire regions, for instance replacing something like TION with NCBE, but not by matching the individual letters. He doesn't need to match letters, or have these regions the same length as those he is replacing. When he comes across a new word ending, he just modifies one of his existing endings slightly and adds it to his list. As such, WLDNCBE and WLDSNCBE might actually translate to very different things. Or he could be doing this syllable by syllable, in which case those two actually do translate to highly similar words. He could be performing this on written english words, but potentially also based on the phonetics. He could even be basing this on a different language if he knows one other than English.