First, there's age: deaths from heart disease tend to happen at old age. In most cultures, death at old age is considered less tragic than the death of an infant or young adult. Also, there's the perceived "naturality" of death: death by natural causes is considered less tragic and less repugnant than murder. And finally there's the fear aspect: a specific murder of A by B for reason X is repugnant but it doesn't necessarily makes other people fear for their lives, while a bombing in a public setting makes people worry that they and their loved ones might be targets too.
I think (i) isn't a very solid argument since we can't estimate the probability of future terrorist attacks the same way we can calculate the probability of winning the lottery, or as we can estimate the probability of x deaths by heart attack in the next 5 years. For example, suppose we count the number of bombings against civilians in pre-2nd-intifada Israel. In 1999 there were 2 bombings. In 2000, when the intifada started, 5. In the following 2 years, 40 then 47. I haven't checked, but I suppose the same pattern would hold for pre- and post-invasion Iraq (note that this isn't an argument pro or anti the invasion).
I'm not saying that the same thing is likely to happen in the US, I'm just pointing out a weakness in the argument. Overall, though, I think Schneier has a point.