After reading the article of Laviolette, Seaman, Barrett, and Woodall, I feel compelled to provide the reader with my own perspective on the topic of fuzzy logic control. For this reason, I have formulated two kinds of comments, the tactical comments, limited in scope to the material covered by Laviolette et al. and aimed at providing clarifications and correcting omissions, and the strategic comments, aimed at providing the reader with a broader viewpoint of this topic. (I just want to clarify that I am not at war with the authors, and the militaristic adjectives used to qualify my comments are simply meant to denote their limited or extended scope.) The authors of the article should be congratulated for presenting an interesting and relatively new topic (fuzzy logic control) to the statistical community. Their presentation, however, lacks the appropriate breadth in fuzzy logic, especially with regard to the selection of the references to be used by the readers. Furthermore, there are a few statements in the article that are not corroborated by the facts. With my tactical comments, I will try to provide useful pointers to interested readers, and I will explain my reasons for disagreeing with some of the authors' statements or with their underlying assumptions. Of course, the authors never intended to provide an extensive treatment of fuzzy logic control (FLC). Therefore, their presentation focuses on the first type of FLC, which was proposed by Mamdani and Assilian (1975) and is typically known in the fuzzy set literature as the Mamdanitype. FLC's have made considerable progress since 1975, however, driven by evolving problem requirements and complexity. With my strategic comments, I will attempt to provide the reader with a broader perspective than the one proposed by the authors. In particular, I will define a framework of comparison for FLC applications and propose a different and, one hopes, more challenging set of questions. For brevity's sake, I will limit my comments to the FLC part of the article. The interested reader however, can find an excellent treatment of the use of fuzzy logic in regression and other statistical techniques in the work of Kacprzyk and Fedrizzi (1992), Terano, Asai, and Sugeno (1992, chap. 4), and Lai and Hwang (1992). 2. TACTICAL COMMENTS