I think you're underrating Colville's defense. He was really the defensive conscious of a fantastic line.
I had Colville last year in the AAA Draft, so I know a good bit about him. I could have sworn I made a bio for him, but looking back I can't find one. From what I remember, that is the only quote I found about his defensive play. He's definitely a good defensive player, but I'm not seeing how his defensive play is as proven or substantiated as Lever or Pederson.
2nd Lines
Both these lines are pretty similar in construction, and are rare lines with playmaking wingers and a goalscoring center. The wings are all post-expansion, so VsX should be a pretty good indicator, and Gagner/Holik played a very similar style.
Prospal: 76, 75, 67, 62, 61, 59, 57, 53, 53, 48, 41(6 year total 400)
Sullivan: 84, 78, 69, 67, 64, 59, 53, 49, 46(6 year total 421)
Semin: 77, 77, 72, 64, 56, 55(6 year total 401)
Pominville: 75, 75, 61, 60, 60, 57, 53(6 year total 388)
So for 6 years, Prospal/Semin are basically the same, and Prospal has 5 more years with scores above 40, so at peak they're basically the same, and Prospal has strong longevity that Semin doesn't have. Semin is noted as being a lackadaisical defensive player, and Prospal isn't Bob Gainey in terms of intangibles, but he's shown an ability to improve his defensive play and physicality as his career has gone on. Prospal holds the advantage here.
In terms of 6 year offense, Sullivan outscores Pominville by 33, and they have an identical seventh season. Sullivan has two more seasons of above 40 scoring, and Pominville doesn't have any, giving Sullivan an advantage in peak and longevity. Both are pretty soft, and have received token Selke support throughout their career. Sullivan received token votes over 3 years, and Pominville over 4 years. Pominville has a slight defensive advantage, but not big enough to make up for a bigger offensive advantage for Sullivan.
For completion's sake, here are their finishes among teammates:
Prospal: 4, 2, 1, 2, 2, 5, 3, 5, 2, 2, 1(7 year avg. 1.71)
Sullivan: 2, 1, 5, 1, 1/1, 2, 4, 2(7 year avg. 1.86)
Semin: 2, 5, 3, 3, 3, 2, 2(7 year avg. 2.86)
Pominville: 4, 2, 2, 3, 3, 1, 2/3(7 year avg. 2.5)
Slashes are when they were traded mid-season and where they would stand for either team given their total point total that season. This re-affirms the offensive advantage for Prospal and Sullivan on the wings.
That brings us to Jaroslav Holik and Gagner. Holik played in a better Czech league than Cerny, so I'll give him that, but his finishes domestically don't seem all that impressive. 1, 2, 2, 3, 6 are the ones listed in his bio, occurring in 66, 67, 69, 72, and 74 respectively. He's also has two strong WC performances to his name in 1969 and 1972. His longevity is pretty poor though. I'm really not sure where he would stand compared to Gagner. Holik seems more of a big brute with weaker skating that was physical with his big size, and Gagner a smaller guy that was faster, and more of an annoying physical guy. I'm not sure how to quantify their respective offensive abilities, but you'd have to think pretty highly of Holik's five seasons of top 6s in the Czech league to take it over Gagner's 10 seasons of useful production in the NHL.
With pretty clear advantages at both wing spots, I think Philadelphia has the advantage in 2nd lines. Center is tough to call with two guys that played a similar style, but in very different circumstances. LWs don't bring much in terms of intangibles, RWs bring a similar amount of defense. Centers are similar tough, two-way players with decent offense.