The next major phase of activity saw 30 enormous Oligocene-Miocene sarsen stones (shown grey on the plan) brought to the site. They may have come from a quarry around 40 kilometres (25 mi) north of Stonehenge, on the Marlborough Downs, or they may have been collected from a “litter” of sarsens on the chalk downs, closer to hand. The stones were dressed and fashioned with mortise and tenon joints before 30 were erected as a 33 metres (110 ft) diameter circle of standing stones, with a ring of 30 lintel stones resting on top. The lintels were fitted to one another using another woodworking method, the tongue and groove joint. Each standing stone was around 4.1 metres (13 ft) high, 2.1 metres (6.9 ft) wide and weighed around 25 tons. Each had clearly been worked with the final effect in mind; the orthostats widen slightly towards the top in order that their perspective remains constant as they rise up from the ground while the lintel stones curve slightly to continue the circular appearance of the earlier monument. The sides of the stones that face inwards are smoother and more finely worked than the sides that face outwards. The average thickness of these stones is 1.1 metres (3.6 ft) and the average distance between them is 1 metre (3.3 ft). A total of 74 stones would have been needed to complete the circle and unless some of the sarsens were removed from the site, it would seem that the ring was left incomplete. Of the lintel stones, they are each around 3.2 metres (10 ft), 1 metre (3.3 ft) wide and 0.8 metres (2.6 ft) thick. The tops of the lintels are 4.9 metres (16 ft) above the ground.

Within this circle stood five trilithons of dressed sarsen stone arranged in a horseshoe shape 13.7 metres (45 ft) across with its open end facing north east. These huge stones, ten uprights and five lintels, weigh up to 50 tons each and were again linked using complex jointing. They are arranged symmetrically; the smallest pair of trilithons were around 6 metres (20 ft) tall, the next pair a little higher and the largest, single trilithon in the south west corner would have been 7.3 metres (24 ft) tall. Only one upright from the Great Trilithon still stands; 6.7 metres (22 ft) is visible and a further 2.4 metres (7.9 ft) is below ground.

The images of a ‘dagger’ and 14 ‘axe-heads’ have been recorded carved on one of the sarsens, known as stone 53. Further axe-head carvings have been seen on the outer faces of stones known as numbers 3, 4, and 5. They are difficult to date but are morphologically similar to later Bronze Age weapons; recent laser scanning work on the carvings supports this interpretation. The pair of trilithons in north east are smallest, measuring around 6 metres (20 ft) in height and the largest is the trilithon in the south west of the horseshoe is almost 7.5 metres (25 ft) tall.

This ambitious phase is radiocarbon dated to between 2600 and 2400 BC. This is slightly before two sets of burials discovered 3 miles (4.8 km) to the west in Amesbury (the Amesbury Archer found in 2002, and the Boscombe Bowmen discovered in 2003) as well as the Stonehenge Archer whose body was discovered in the outer ditch of the monument in 1978.

At a similar time a large timber circle and another avenue were constructed overlooking the River Avon 2 miles away at Durrington Walls. Opposing the solar alignments at Stonehenge, the circle was orientated towards the rising sun on the midwinter solstice, whilst the Avenue led from the river to the circle on an alignment to the setting sun on the summer solstice. Evidence of huge fires on the banks of the Avon between the two avenues also suggests that both circles were linked, and perhaps formed a procession route used on the longest and shortest days of the year. Parker Pearson speculates that the wooden circle at Durrington Walls was the centre of a ‘land of the living’, whilst the stone circle represented a ‘land of the dead’. The Avon would have served as a journey between the two.