In this paper, we achieved two aims. Firstly, laser light-scattering intensities in methyl red (MR) azo and disperse red (DR) anthraquinone dye-doped nematic liquid crystal (NLC) molecules were measured versus scattering angle and applied bias voltage. The following three NLC molecular structure parameters were determined: the visco-elastic constant ratios K11/K 22 and K33/K22 by data-regression and Freedericksz voltages from the graphs drawn. All these NLC parameters were found to be dependent on the kind of the dye used. As the second aim, by nonlinear universal function approximator layered feedforward neural network (LFNN), we constructed explicit form of empirical physical formulas (EPFs) for theoretically unknown nonlinear azo and anthraquinone dye-doped NLC scattering intensity functions. Excellent LFNN test set (i.e. yet-to-be measured experimental data) predictions prove that the constructed LFNN-EPFs estimate unknown intensity functions consistently. The LFNN-EPFs, too, confirmed the dependency on the kind of dye used. In conclusion, physical laws embedded in the scattering data can be consistently extracted by LFNN. One significant potential application in molecular nonlinear optics domain is that these LFNN-EPFs, by various mathematical tools such as differentiation, integration, and minimization, can be used to obtain further NLC scattering intensity knowledge related molecular structural parameters. Such knowledge in turn may prove useful in developing new optical materials. © 2011 Elsevier B.V. All rights reserved.