A brief introduction to the brain--mind problem leads on to a survey of the neuronal structure of the cerebral cortex. It is proposed that the basic receptive units are the bundles or clusters of apical dendrites of the pyramidal cells of laminae V and III-II as described by Fleischhauer and Peters and their associates. There are up to 100 apical dendrites in these receptive units, named dendrons. Each dendron would have an input of up to 100 000 spine synapses. There are about 40 million dendrons in the human cerebral cortex. A study of the influence of mental events on the brain leads to the hypothesis that all mental events, the whole of the World 2 of Popper, are composed of mental units, each carrying its own characteristic mental experience. It is further proposed that each mental unit, named psychon, is uniquely linked to a dendron. So the mind--brain problem reduces to the interaction between a dendron and its psychon for all the 40 million linked units. In my 1986 paper (Proc. R. Soc. Lond. B 227, 411-428) on the mind--brain problem, there was developed the concept that the operation of the synaptic microsites involved displacement of particles so small that they were within range of the uncertainty principle of Heisenberg. The psychon--dendron interaction provides a much improved basis for effective selection by a process analogous to a quantal probability field. In the fully developed hypothesis psychons act on dendrons in the whole world of conscious experiences and dendrons act on psychons in all perceptions and memories. It is shown how these interactions involve no violation of the conservation laws. There are great potentialities of these unitary concepts, for example as an explanation of the global nature of a visual experience from moment to moment. It would seem that there can be psychons not linked to dendrons, but only to other psychons, creating what we may call a psychon world.