The best generation for the evaluation and selection of a trait depends on its polygenic level and heritability. Little attention has been paid to integrating, in the same rice recurrent selection (RS) scheme, the separation of breeding generations. The option we propose is a new parallel and interlaced RS (PAIRS) scheme. We evaluated and selected earliness and grain size in the S0 generation, complete resistance to leaf blast in S1 plants, partial resistance to leaf and neck blast, earliness and yield components in S2 lines, gelatinization temperature, estimated by the extent of alkali spreading in S3 grains, and tolerance to rice delphacid (Tagasodes orizicolus) mechanical injuries in S3 plants. Although this innovative scheme may seem complicated overall, this first application of PAIRS highlights that the improvement of every methodological step is very easy. Through a complete recurrent selection and genetic recombination cycle, we obtained good progress for grain length, partial resistance to leaf blast and gelatinization temperature (alkali spreading value), and excellent progress for targeted earliness, complete resistance to leaf blast, partial resistance to neck blast and yield. The population size and the number of plants or lines chosen were adequate and we can recommend them to partners.