Meanwhile, I have got hold of the famous Parnas paper "On the Criteria To Be Used in Decomposing Systems into Modules" and I hope to get hold of his book "Software Fundamentals" quite soon (or at least some of the essays in the book; the book appears to cost a bit more than I can comfortably budget at this point.).
I cannot claim that I've fully understood the Parnas paper (particularly in regard to its implications in the specific context of my interests in 'system design' of 'non-computer systems'). I do observe that it does seem to be generating a fair amount of controvery even today (though it was originally published around forty years ago!) - I do hope I shall be able to get most of my doubts clarified in due course. I observe that, outside the field of computer systems, the 'decomposition of a system into its component modules' should really pose no great difficulty to anyone who has adequately understood how to apply simple tools for the analysis (and synthesis) of systems.
It does strike me that at least some of the controversy about the paper is occasioned by inadequate understanding of its arguments and conclusions. (Apparently, I had almost forgotten that this is actually a pretty widespread phenomenon in human discourse - not one confined to our debates here at Math-teach!)