The Fuzzy Teaching Machine originating in work by J. J. Buckley and colleagues is implemented so that its inherent choices can be exercised and some of its options explored. The ``machine" links fuzzy and neural net paradigms for computing in words for control. Comparisons are made to models such as the MATLAB Mamdani and Sugeno Fuzzy Information Systems, and our own rule processing system. Modularity is cited for: educing answers to problems, simplifying implementations, and integrated solutions involving the FTM, our rule processing system and other systems. Implementation details play a cameo role in the discussion. An application is made to a model for humans (usually children) and robots (hence `RoboKid") on development and use of external memory strategies in carrying out tasks relating to movement of objects in a laboratory setting.