The paper is concerned with on-line process steady-state optimization under uncertainty. In such cases a single process model optimization can yield a set-point far away from the one optimal for the true process. The way to improve the set-point is to apply steady-state feedback, i.e., an iterative optimizing algorithm utilizing new measurements available after every subsequent set-point application. Integrated System Optimization and Parameter Estimation (ISOPE) method yields subsequent set-points converging to the true process optimum, despite uncertainty. It requires, at every iteration, model parameters to be updated under certain additional equality constraint. The aim of the paper is to present how the ISOPE can be redesigned resulting in a new structure without this constraint. Moreover, the parameter estimation itself is then not necessary at every iteration, although possible when reasonable.