BACKGROUND: Children with learning disabilities are a heterogeneous group of children with a common characteristic discrepancy on the progress and development of their individual learning abilities. A few statistical analyses have been published regarding the factor analysis of the Greek Edition of Wechsler Intelligence Scale for Children-III. The aim of the research is the emergence of a new factorial model which describes the General Intelligence (g) of children and adolescents with learning disabilities, and that differs from the already existing intelligence models. This study aims to compare three-factor structure models of WISC-III in children with learning disabilities in the Greek population. METHODS: A sample of 50 children were selected on the basis of research criteria from a total of 122 children who evaluated in a child psychiatric service in a general hospital, in a residential area in Greece. The Wechsler Intelligence Scale for Children-Third Edition was used to assess children's cognitive function. Using multi-factor analysis, three alternative factor models were compared. RESULTS: Analysis of factor structure models suggests a new bi-factorial model that more appropriately describes the areas of cognitive development of children with learning disabilities. The first factor includes Comprehension, Picture Arrangement, Coding, Block Design, and Object Assembly, whereas the second one combines Information, Similarities, Arithmetic, Vocabulary, and Picture Arrangement. CONCLUSIONS: The present study shows the existence of a factorial model with two factors: one aggregating the Comprehension verbal subtest with four performance subtests and the other the Picture Arrangement performance subtest with four verbal subtests. This two-factor model includes the loadings in two factors that relate to sequencing abilities and verbal reasoning abilities of children. These findings assert the clinical utility of the intelligence evaluation in the specific population.