This paper addresses working with MCDM methods unable to model correctly a scenario. Most Decisions Makers (DM) are forced to select a method even knowing that it does not fulfill it, while others are con- scious that they are simplifying the scenario. The result for the abridged version or proxy of the real scenario may be correct, but it does not repre- sent the real one, and consequently, the extracted conclusions do not per- tain to the original problem. Since no mathematical formula exists to solve problems of this nature, and because results cannot be validated, it is easy to state, as many authors do, that the problem was successfully solved us- ing a particular method, although obviously no consideration is given to the true nature of the problem, which is usually easily seen by reading its characteristics and by examining the process used to solve it. In addition, this paper sustains that an MCDM problem is a system of linear equations, and as such, everything must be accounted for and solved at the same time; this concept is overlooked by most MCDM methods. Some empha- sise partitioning a complex problem and solving each part separately, which is beneficial because it facilitates the comprehension of the prob- lem, but it does not mean that the problem is solved in its integrity. Usual- ly, the whole is something more than the sum of the parts. Keywords: MCDM, alternatives, criteria, modelling, networks, SIMUS.