The distribution of the triplet to amino acid correspondences in the genetic code matrix contains blocks of similarity. There are (a) groups of similar triplets coding for the same amino acid, which is called code degeneracy, and (b) clusters of similar amino acids corresponding to similar triplets. Processes that led to this regionalization have been investigated through a variety of perspectives but no consensus has been reached and no model has been convincing enough to drive experimental tests. Most traditional has been the hypothesis that the code was derived from the standard evolutionary processes of testing variations in the correspondences through the fitness measure of reaching distributions in the matrix space in an optimal manner so that the effects of mutations on protein phenotypes would be minimized, that is, with reduction of the intensity or of the deviant quality of the functional alterations associated with variations. In contrast, the self-referential model for the formation of the code is based on an original regionalization of characters through the concerted superposition of the two components of the encodings: the four modules of dimers of tRNAs are occupied sequentially by sets of amino acids that are also sequentially devoted to fulfilling specific functions in the protein sites and motifs to which they preferentially belong. Therewith, part (b) of the error-minimizing property follows. Part (a) of the property, the code degeneracy, is derived from the synthetase character of developing specificities directed initially to the principal dinucleotides of the triplets, resulting in tetracodonic degeneracy. This was later partly modified during evolution according to the developments of codon usage and the introduction of new amino acids.