The problem of the functional structures research is considered in this example of information systems. A feature of such research is that it is not always possible to ensure that the research results will match reality. This is a topic of current interest in the field of design and analysis of information security systems and software analysis for undeclared capabilities of systems in general. By undeclared capabilities, we refer to a functionality available in software that is invisible to users and can be used / exploited by an intruder. This paper presents a model of a researcher and of a functional object investigated by him. Based on this model, informational limitations of the researcher are shown. The mathematical model of the subjective structure of an investigated system is constructed. It is shown in which cases this structure is stable. This article answers the question of if the researcher can claim that his subjective functional structure corresponds to the actual structure of the investigated system. We provide examples of such approach on certain mathematical models of information security.