The eccentricity of Earth's orbit doesn't make for more or less extreme summer/winter differences, but more or less extreme northern vs. southern hemisphere winter/summer differences. This is central to the theory of how the Milankovitch cycles are tied to glacial and interglacial periods of the current ice age.

What makes seasons less extreme is a closer-to-right-angle angle of the axis of Earth's rotation to the ecliptic plane. This angle does vary a fair bit over time.

In geologic time there are four astronomic parameters of Earth's movement, and one for the Sun's, that can affect the climate and life:

     - eccentricity of Earth's orbit
     - precession of the equinoxes
     - angle of the axis of Earth's rotation to the ecliptic
     - distance from the Earth to the Sun (which, besides
       eccentricity, varies over long geologic time due to
       interactions with Jupiter and other planets)

When the northern hemisphere's summer coincides with the perigee of Earth's orbit around the Sun, insolation is at its highest over the most landmass, thus ensuring that ice does not accumulate on land over the summer and that any existing ice melts. This is likely the cause of the start of interglacials, while the opposite situation is likely the cause of the start of glacial periods. The angle of the axis of Earth's rotation to the ecliptic also plays a role, in this, obviously, since the smaller the angle the more extreme the seasonal differences. So goes the theory anyways.

And, of course, the galactic neighborhood traversed by the Sun can make a difference. We don't want to be near stars about to go supernova...