Makis,
I got my data for earthquakes in Greece from the following two
sites in the absence at the time of data on the NOA site:
http://seismo.ethz.ch/redpuma/redpuma_ami_list.html
and
http://www.emsc-csem.org/cgi-bin/ALERT_all_messages.sh?1

Please check these sites and you will find the earthquakes
I noted in the earlier post.

I would be the first to agree that local data on small events
is usually more correct, but in the absence of that data, it
is necessary to use teleseismic or quick determinations temporarily.

Please note that the second event listed gave the time in PDT
(Pacific Daylight Time - - Don's timeframe). It is the same
as the event of September 2 at 06:59 which you did find.

The reason for using PDT here is to note that in Don's time frame (PDT)
this earthquake occurred 1 minute before his window closed. Of course,
the event really occurred outside Don's time window in both UT and
Local Time.

I will cede to your knowledge of Greek quakes and concede that
the MEDNET location for the second is incorrect. They gave:

2001/09/02 06:59:05.5 37.4N 21.5E 10 Mb5.5 A MIX SOUTHERN GREECE

2001/09/02 06:59:04.4 37.3N 21.7E 10G Ml4.7 A INGV SOUTHERN GREECE

2001/09/02 06:59:01.8 36.9N 21.5E Mb5.4 A SED SOUTHERN GREECE
Source:
http://www.emsc-csem.org/cgi-bin/ALERT_all_messages.sh?1

Usually three independent determinations will give a location near
the correct, but apparently not in this case.