The mechanisms of the photochemical isomerisation reactions are investigated theoretically using the model systems, 1,2-dihydro-1,2-phosphaborine (5) and 1,2-dihydro-1,2-alumazaine (6), using the CAS(6,6)/6-311G(d,p) and MP2-CAS-(6,6)/6-311++G(3df,3pd)//CAS(6,6)/6-311G(d,p) methods. For each model reactant, three reaction pathways, which lead to three kinds of photo-isomers, are examined. The structures of the conical intersections, which play a key role in such photo-rearrangements, are determined. The thermal (or dark) reactions of the reactant species are also examined, using the same level of theory, to provide a qualitative explanation of the reaction pathways. These model investigations demonstrate that the preferred reaction route for these two aromatic heterocyclics is as follows: reactant -> Franck-Condon region -> conical intersection -> photoproduct. The theoretical evidences anticipate that after irradiation of 5, the photoproduct yield of the Dewar BP-isomer, 8, should be larger than that of the Dewar BP-isomer, 7, whereas no Dewar BP-isomer 9 can be observed. Moreover, the present theoretical data predict after irradiation of 6, all three Dewar AlN-isomers (10, 11, and 12) and the starting molecule, 6, are produced.