Hi Pál,
While I can see where Ken is coming from with regard to image number 3, what looks like a very long primary projection is, in fact, due to a sitting gull in the background. The same darker-mantled bird present in the foreground, and in all of the other images, is tricky from these images alone, though perhaps someone with more varied experience of the other taxa can come along and correct me, or offer a more definite identification.
I more or less suspect that identification comes down to the three taxa that you outlined in your last post, but, for me at least, it is very tricky to tell for certain which taxon is most likely. If I saw this in Ireland, I'd probably overlook it as a graellsii, and perhaps this is what it is, but, in Romania, other possibilities need to be considered. It could be argued that having a white mirror on P10 but none on P9 would count against graellsii, but some birds can lack this, it wouldn't be that unusual in and of itself. On the other hand, and I stress that my experience of armenicus and heuglini is extremely limited, it might be expected that adults of both taxa would be more likely to have black all the way in to P4, and possibly even P3, though the subject bird has black only to P5, from what can be seen. In addition, the lack of primary moult would be another point against the late-moulting heuglini, at least when compared to the majority of birds, and, if one is considering armenicus, this would have to be one of the rare adults without black bill markings. Another possibility, I suppose, is that this might be some sort of hybrid from the overlap zone between Armenian and Yellow-legged Gulls, but I lack much knowledge of what these could look like. It does seem too dark above for michahellis, which most of the accompanying gulls seem to be.
Hopefully someone like Peter or Lou will chime in with their own far more educated opinions!
Regards,
Harry