Seed quality is a complex trait that is the result of a large variety of developmental processes. The molecular-genetic dissection of these seed processes and their relationship with seed and seedling phenotypes will allow the identification of the regulatory genes and signalling pathways involved and, thus, provide the means to predict and enhance seed quality. Natural variation for seed-quality aspects found in recombinant inbred line (RIL) populations is a great resource to help unravel the complex networks involved in the acquisition of seed quality. Besides extensive phenotyping, RILs can also be profiled by -omics technologies, such as transcriptomics, proteomics and metabolomics in a sophisticated so-called generalized genetical genomics approach. This combined use of physiology, genetics and several -omics technologies, followed by advanced data analysis, allows the construction of regulatory networks involved in the various attributes of seed and seedling quality. This type of analysis of the genetic variation in RIL populations in combination with genome-wide association (GWA) studies will allow a relatively rapid identification of genes that are responsible for quality-related traits of seeds and seedlings. New developments in several -omics technologies, especially the fast-evolving next-generation sequencing techniques, will make a similar system-wide approach more applicable to non-model species in the near future and this will be a huge boost for the potential to breed for seed quality.