Galaxy models for radiative rransfer I have addressed the validity and the limitations of the axisymmetric smooth exponential model used for the description of spiral galaxies. I have quantified the error that is introduced in the parameters describing a galaxy when the spiral structure is neglected and I have found that errors are neither systematic nor very large. Therefore I have concluded that an axisymmetric model constitutes a valid description of real spiral galaxies. On the other hand, I have found that if the clumpy distribution of dust is not taken into account, then the amount of dust may be underestimated by at most 40%. Spectral energy distribution of spiral galaxies I have constructed a Spectral Energy Distribution (SED) model for spiral galaxies taking into account the relative spatial distributions of the stars and the dust, the locally absorbed ultraviolet (UV) radiation in star forming regions and the transiently heated dust grains. I have applied it to the edge-on spiral galaxies NGC 891, NGC 5907, NGC 4013, UGC 2048 and UGC 1082 using data from the literature. The fitting parameters of the model are the star formation rate (SFR), the dust mass and the proportion of the locally absorbed UV radiation. The SFRs obtained for these galaxies are in excellent agreement with the values of SFR found in the literature. The inferred dust masses, on the other hand, suggest that the dust masses quoted in the literature constitute an underestimation of the actual dust content in these galaxies. I also find that in normal galaxies (contrary to starburst galaxies) a large percentage of the dust heating is due to the old stellar population. N-Body simulations of barred galaxies I have investigated the morphology and kinematics of barred galaxies from the theoretical point of view analyzing N-body simulations representing barred galaxies. I have run models with different halo and bulge distributions in order to investigate the effect of the central concentration of the halo and the bulge on the morphological properties of the bar. I have found that a massive halo in the inner parts of the model galaxy results in a more elongated, more peanut-like and more strong bar. Several other morphological and kinematic properties are also investigated and a case for two types of bars is made. Suggestions for future studies on SED of galaxies So far, the limited data at wavelengths longer that 100 microns have allowed the application of SED models to a small sample of normal (non starburst) galaxies. In the past, statistical studies have suffered from over-simplified models leading to contradictory results about the distribution of dust in galaxies. As new data at FIR wavelengths are obtained for more galaxies, it is essential to develop a standard technique for the interpretation of these data. I think that on the basis of my study, a standard technique can be developed and tested on the very well studied cases I have presented. Then this technique can be applied to a much larger sample of observations of galaxies which will become available. Face-on spiral galaxies should be included in this study, as well as ellipticals and even irregulars. Suggestions for future studies on N-body simulations of barred galaxies While there have been a lot of studies on the subject of stellar orbits in barred galaxies, all these studies deal with analytical potentials. The analysis of the orbits in ``live'' potentials (i.e. potentials derived from N-body simulations) will reveal the mechanisms responsible for processes such as the formation of rings, the transfer of angular momentum from the bar to the disk and the halo, and the trapping of stars in resonances. Understanding the mechanisms that drive the evolution of galaxies from the theoretical point of view, is an indispensable step for the interpretation of observations of barred galaxies in the local as well as in the distant universe.