Despite the considerable progress in the theories of stellar evolution and nucleosynthesis, important questions related to the production sites and stellar yields of some of the CNO isotopes still remain unanswered. Single low-, intermediate- and high-mass stars as well as binary stars can produce CNO elements. These elements are then restored into the interstellar medium on different time scales, according to the different nature of the stellar progenitors. In this work, we use a detailed chemical evolution model for the Galactic halo and disc to shed light on the contributions to the abundances of the CNO group nuclei provided by different stellar factories during the whole Galactic lifetime. N, We suggest that the yields of the secondary isotopes require major revisions. We also show that, if novae actually contribute important amounts of the secondary isotopes C-13; N-15 and O-17, as claimed by several authors, the behaviour of the C-12/C-13, N-14/N-15 and O-16/O-17 ratios in the Milky Way is better explained. In particular, our chemical evolution model results are in agreement with recent claims that novae should be the major sources of N-15 in a galaxy, while massive rotating stars should not produce N-15, at a galactic level. However, in our opinion current nova yields should be revisited. In particular, it would be worthwhile to study the dependence of nova yields on metallicity.