IMHO and from my experience, the main keys to addressing complexity of software are just to (A) define the data, input, inside the software, data base, key-valu

IMHO and from my experience, the main keys to addressing complexity of software are just to (A) define the data, input, inside the software, data base, key-value store, etc., and output, (B) organize the work for any real time parts, (C) organize the code essentially a calling hierarchy of pieces, where each piece is fairly easy to test well, and where the purpose of each piece is fairly easy to document, and (D), most important, document everything so that it is all easy enough to understand, check, modify, etc. For really big projects, there is more, but (A)-(D) are the basics and where still in practice a lot of problems are. A curious but crucial part is that nearly all the meaning of the system is just in the documentation, not the code.