Artificial Neural Networks (ANNs) are an advanced modelling technique applied in numerous scientific fields, including limnology. This study explores the application of the Kohonen Self-organizing Map (SOM), which is an unsupervised ANN, utilized to explore the interactions of water quality parameters and how the associated environmental data samples are grouped into clusters with different environmental properties. Therefore, the SOMs can be used for assessing the overall status of lakes. Regarding eutrophication phenomenon, which is a major issue for Mediterranean lakes, the SOMs have been proved to be an excellent modelling tool for examining the trophic status of lakes, superior to many other modelling techniques (e.g., Cluster Analysis). This is because the SOMs can provide insight into complex limnological data, but also to analyze multidimensional environmental data and simplify them into visual information with the use of the SOM’s Component Planes. Taking SOM-based modelling studies of water quality as examples and their practical implementation, this study provides a guideline on how to develop a SOM model and to analyze the outputs. Additionally, the role of SOM models as management tools for lakes is pointed out and gives directions for future limnological research.