Yes, you are quite correct. The approach I gave is an approximation. In a laboratory, my method is probably too general for most applications. In the oilfield, this is accurate enough (commonly referred to as OFCE, or Oilfield Close-Enough).
As an example in the oilfield, a blend (methanol and water) in a tank quite often contains a high level of contamination (1% crude oil, probably sand, clay and other possible contamination). Most calcuations such as this frequently would like to know what the freeze point is, or the affinity to absorb more water.
Again, you are right when talking about chemistry within, lets say, pharamceutical areas, in the oilfield, this is accurate enough.
Thanks for the comments.