Recent evidence derived from transgenic mouse models and findings in humans with mutations affecting thyroid hormone (TH) metabolism have convincingly supported a model of TH signalling in which regulated local adjustment of active TH concentrations is far more important than circulating plasma hormone levels. Although this theory was put forward several years ago and has been supported by significant, but inherently indirect evidence, recent insights from targeted deletion of the genes encoding deiodinase (Dio) isozymes have revived this model and greatly increased our understanding of TH metabolism. However, gene targeting proved to be a double edged sword, since the overall model was supported, but several predictions are apparently not consistent with the new experimental evidence. Human genetics further provided additional exciting data on the physiological role of Dio isozymes that need to be incorporated into any model of TH biology. The recent identification of mutations in the T3 plasma membrane transporter MCT8 has sparked new interest in the role of TH in brain function, since affected patients suffered from psychomotor retardation. Moreover, selenium (Se) and TH physiology have finally been unequivocally connected by newly identified inherited defects in a gene involved in selenoprotein biosynthesis. Finally, a link between Dio expression and energy metabolism has been delineated in mice that may hold great promise for the management of the adiposity pandemic.