By their textbook definition, RBMs are probabilistic graphical models, which—given what we've already covered regarding the structure of neural networks—simply means a bunch of neurons that have weighted connections to another bunch of neurons.
These networks have two layers: a visible layer and a hidden layer. A visible layer is a layer into which you feed the data, and a hidden layer is a layer that isn't exposed to your data directly, but has to develop a meaningful representation of it for the task at hand. These tasks include dimensionality reduction, collaborative filtering, binary classification, and others. The restricted means that the connections are not lateral (that is, between nodes of the same layer), but ...
