Fuzzy modeling is one of the techniques that is used for modeling of nonlinear, uncertain, and complex systems. One of the aspects that distinguish fuzzy modeling from other black-box approaches like neural nets is that fuzzy models can be transparent to interpretation and analysis to a degree. However, the transparency of a fuzzy model is not achieved automatically, and needs careful consideration during the learning phase from the available data. In this presentation, we consider several methods for learning interpretable fuzzy models from data. Interpretability is posed as additional constraining of the model parameters. Some practical examples are also given.