Fellow Wikipedian hdamm is currently creating vector maps for all the districts superseding the pixel maps I created years ago. While doing so he ran into similar cases of the contradicting maps like those I lamented about a year ago, most strikingly in Kalasin.
The first and most notable discrepancy between his and my maps resolved quite easily - in the district Kamalasai a quite large bulge in the eastern boundary showed up, which turned out to be the subdistrict Nong Paen. For whatever reason this subdistrict was not added to the minor district Khong Chai when it was created in 1997, yet my map source did show belonging to Khong Chai.
Much more tricky was the second one, an additional bulge in the southern boundary of Tha Kantho, to be exact some area of Nong Yai subdistrict, Nong Kung Si district being added to Kung Kao subdistrict. Half of the maps showed the bulge, the other half did not. For example the map at amphoe.com show the bulge, whereas Google Earth or a map from the provincial statistics office don't have it. Nothing in the Royal Gazette suggests a change in the boundary at that area - a recent boundary adjustment would be a nice explanation, but there apparently was none.
But finally the Gazette came to my rescue - an 2008 announcement declaring some part of Tha Kantho as land reform area includes a map delineating the boundary of the affected areas, and this includes the boundaries of districts and subdistricts. And there it shows the bulge, same as in the source hdamm used.
area in terrain mode one can see that the boundary roughly follows the boundary of the plateau, though mostly covered by forest.
And now I found this alternative and I believe authoritative source of maps I tried to check those two cases from last years posting, and it confirms the original map of Satun without the large part of Trang, whereas for Lopburi the district Lam Sonthi does not extend as far north as it does in some maps.