In the present study, after doing some numerical modeling of the problem with the finite volume method and specifying the effective parameters, by extracting a mathematical response surface , the effect of adding a microtube on the thermal and hydrodynamic performance of a wavy micro heatsink (MCHS) is investigated. The interaction between the input parameters of the problem, which are the geometrical variables of the microtube and their effect on the output parameters such as total heat transfer coefficient (h), total pumping power, total thermal resistance, entropy generation (sgen), etc. are evaluated by an RSM statistical model. In the present study, the microtube and the microchannel are considered simultaneously in a wavy MCHS and the new geometry of MCHS for cooling Central processing units is presented. The results show that increasing Reynolds number (Re) and nanoparticle concentration (ϕ) in all studied geometries, improves MCHS performance and reduces CPU surface temperature as well as thermal and total irreversibility rate and causes temperature uniformity however, increasing the Reynolds number has a negative effect on the pumping power of the system. Then, the solution of the genetic iterative algorithm was used to find the best type of Response Surface for each output parameter. And then, the accuracy of mathematical models made by RSM for objective functions was proved. Other output parameters considered are microchannel and microtube pressure drop, mean and maximum CPU surface temperature, Re of microchannel and microtube, which according to design parameters include system flow rate, ratio microtube flow to the microchannel, microtube diameter and microtube distance from MCHS surface, response surface coefficients are obtained. It is found that the accuracy of the Response Surface of the temperature uniformity on the CPU surface has a very high error and as a result, this Response Surface does not provide a good estimate.