An approach to deal with increasing complexity in control tasks is used the human mind as an archetype for a reasoning unit. In this article, psychoanalysis is used as theoretical framework for the described model. The main elements are Id, Ego, and Super-Ego. This article deals with fail-safe aspects of such a complex architecture. What will happen if one or even two of the main elements fail to operate. The technical architecture presented divides the model into two parts-one fail-safe part with the Id and selected elements of the Ego and the other one with the rest of the Ego and the Super-Ego. The advantage in doing so is that while the latter one is able to perform complex tasks, the first one guarantees that-in case of an error-the system will not fail totally.