http://www.fig.net/resources/proceedings/fig_proceedings/fig2007/papers/ts_4c/ts04c_03_hormoz_ziari_1504.pdf

http://people.ku.edu/~gbohling/cpe940/Kriging.pdf

http://www1.mate.polimi.it/~bonavent/Homepage_Luca_file/lezkrig005.pdf

see this paper

https://www.researchgate.net/publication/317042483_Studying_Malaria_Epidemic_for_Vulnerability_Zones_Multi-Criteria_Approach_of_Geospatial_Tools

Article Studying Malaria Epidemic for Vulnerability Zones: Multi-Cri...

These links are not fro ArcGIS, but they are also very useful for answering your question:

https://support.goldensoftware.com/hc/en-us/articles/231348728-A-Basic-Understanding-of-Surfer-Gridding-Methods-Part-1

https://support.goldensoftware.com/hc/en-us/articles/230756107-A-Basic-Understanding-of-Surfer-Gridding-Methods-Part-2

https://support.goldensoftware.com/hc/en-us/articles/226661908-What-Surfer-gridding-method-is-best-for-my-data-file-

https://support.goldensoftware.com/hc/en-us/articles/230756747-What-Surfer-gridding-method-should-I-use-Gridding-for-non-geostatisticians

Please also read the following papers:

Article The variogram sill and the sample variance

Article A tutorial guide to geostatistics: Computing and modelling v...

You're talking about interpolation.  Fundamentally, you need to be sure that the interpolation method you are applying matches the nature of the data you are applying it to.  How does the data change with distance?  Does the interpolator capture that change?  Kriging is flexible in this regard because you tailor its function to match the data.  To use it effectively, you need to be able to understand the tool's function - research is the only way to do it in the first instance.  

Thanks so much Laszlo your answer is very helpful

 Many thans to Firoz, Jin and Christian for sharing your answer, the files are realy valuable and will increase my experience in that point

In order to get more accurate results, I suggest that you divide your surface layer into sub-layers based on homogeneity, then apply the IDW  for each sub-layer separately, then combine them again.

Also, the Geostatistical Analysis tool allows you to create a random subset of your data points.  (validation data)  The rationale of this is to interpolate using the interpolator of choice and use the subset to validate the surface, which quantitatively measures the surface compared with the data that was left out of the interpolation.  This can be a very powerful tool because it allows you to repeatedly interpolate with different parameters and then decide on the interpolator and parameters that generate the most accurate surface.  The final interpolation is then performed on the entire dataset.