Interpretability has become an essential concern for Machine Learning (ML) models in critical areas such as healthcare, law and manufacturing industries. Fuzzy Pattern Trees (FPTs) are tree-based structures in which the internal nodes are fuzzy operators, and the leaves are fuzzy features. Representing input data using fuzzy logic usually brings more interpretability to the features since it separates the data into specific, often meaningful, parts of their domain, usually associated with a descriptive term. This work uses Genetic Programming (GP) to evolve FPTs and assess their performance on 20 benchmark classification problems. Furthermore, we experiment using Lexicase Selection with FPTs and demonstrate that selection methods based on aggregate fitness, such as Tournament Selection, produce more accurate models before analysing why this is the case. We also examine results from the perspective of the interpretability of the models, propose new parsimony pressure methods embedded in Lexicase Selection, and analyse their ability to reduce the size of the solutions. The results show that for most problems, at least one method could reduce the size significantly while keeping a similar accuracy.