Modern Y DNA distributions don't correlate 1:1 with past admixtures because in the meantime you can have founder effects, further gene flow from neighbouring regions who are autosomally similar but might have a different Y structure, etc.
On the Germanic question, one explanation could simply be that the Germanic groups that settled into Italy had already admixed with other peoples in the Balkans and/or the Alpine region and would not be very rich in I1. And clearly some of the people who moved from modern Austria or Switzerland would also carry R-U152 and other haplogroups that were already present in Iron Age Italians (J-L283, G-L497 for example). So you can't automatically assume all of these haplos in modern Italians are "native" since the Iron Age.
On the "Middle eastern sources", fist of all we don't have to imagine a direct emigration from the Middle East. Depending on the time period, a lot of these people would have come from Rome or other Imperial centers in Italy. And in the Late Antiquity, people with partly Near Eastern would be just Romans, without any quotes.
But anyway, if we want to do a ballpark estimate of the the impact of the "East Med" cluster on the Y frequencies, you'd have to add up various branches of J2a and G2a (J-L70, J-M92, G-M406, G-L13 and others), most of haplogroups J1, L and T, E-M84, some E-V13 (since we've seen by now a number of samples carrying E-V13 but having the classic Cypriot/Dodecanesian profile), R-Z93. For R1b, not all of it should be automatically assumed to be of recent European origin, for exaple if you go to
R-Z2103 on yfull and CTRL+F for 'ITA', you'll see that a big chunk is of BA Caucasian or Middle Eastern origin.