I don’t know about the Hungarian-Turkish connection; someone more versed in their language histories would have to weigh in on that. I think it’s worth mentioning that some linguists group Uralic (Hungarian) and Altaic (Turkish) into one large family called Uralic-Altaic.
Right, not just in Sanskrit but in Indo-European in general. But the source I mentioned earlier, and other sources I’ve looked at, seem to indicate that nobody is really sure whether the agent suffix and the kinship suffix really are related, or if not, why they’re so similar.
Just to point out that although this is a good reference and MacDonell is in general reliable, he wrote a long time ago and there have been many breakthroughs in I-E Studies that he could not have known about. Doesn’t mean he’s wrong, just not the horse’s mouth for IE stuff.