Once an element is developed as a model, the model can be changed. A changed model will have other associated truths, so that, when treated as an element of the main model, it has another set of properties. Thus the properties cannot be changed in a direct way, but through the changes in the model. In any case, a main model can operate only with elements, regardless of the fact that the element is or not already developed as a model. We already use terms as "long range models" or "short range models". Let's define them. A long-range model has already been defined as a model with its own elements developed as models. But here we will prefer another alternative definition. A long-range model is a model which reaches its aims by activation and deactivation of some of its elements. Such elements are already developed as models. A short-range model reaches its aims by direct activation. Example: to switch on the light in a room, a ZM model will make a ZAM. That ZAM will simulate the action. Based on this simulation it will activate an AZM which, in turn, will switch on the light. The ZM-model will confirm the success of the activity of the short-range ZAM model. Example: To travel from a place to another, a ZM will make a ZAM. The ZAM will make some ZAMs. These ZAMs will make some others ZAMs. For any specific activity there will be a ZAM. Once a ZAM has reached its aim, it will be deactivated by the ZAM-model which activated it, and a new ZAM is activated. The general control belongs to the main-ZAM. The main-ZAM can be modified by the main ZM. Long-range models do such activity. Example: we enter a room and switch on the light. The light really switches on. A local-ZM gets this information based on IR. But, a long-range ZM, which contains the local-ZM as an element, understands that the light had been broken, and now it is on. The local-ZM acts here as a shorter-range model. It does not understand the general environment. The main-ZM (which contains the local-ZM as element) is a long-range model. We already saw that any model can activate any of its elements to self-develop as a model. Once an element is activated, it develops itself as a model. In turn, this new model can also activate as a model any of its elements. This "depth" has only technological limits. In fact, almost any model of the brain is a long-range model. The definitions, which are already given, respect this feature. Now we shall present a list with the main models of a brain: ZAM: these models are usually long-range models. A ZM model makes them. Their main function is to change the external reality. They are made for immediate activation (to drink water from a glass, for instance) or they could be models that give the orientation of the whole life of a person or anything between these very large limits.