Neurofuzzy systems are computing architectures whose main features arise as an effect of an important synergy occurring between two fundamental facets of information processing such as fuzzy computing and neurocomputing. Their underpinning is in a complementary character of fuzzy sets and neural networks. The latter is oriented toward more numeric processing of massive data. On the other hand, fuzzy sets tend to address an important aspect of processing focused on the formation of information granules and their processing. We discuss various architectures and learning paradigms concentrating of many ways in which such essential symbiotic links between fuzzy sets and neurocomputing are established and exploited.