I'm not sure that I have a full enough understanding to explain it, but I'll try.
The first thing is that you can ignore high intermediate values if subsequent values are lower. So your hops 3 and 4 have high values, but there are much lower values after them. All this means is that those servers prioritise actual through traffic and give low priority to ICMP (ping) responses. All that matters is how fast the data passes through them.
But hop 2 is just about the lowest value of the lot, and it's still rather on the high side. To give you a comparison, when I run a traceroute the second hop (my ISP) consistently shows values around 20 ms, and a ping to bbc.co.uk is generally less that 25 ms (and my connection is interleaved). So although I can't be certain, I suspect that your ISP is causing a bit of extra latency.